Obstacle Avoidance Performance Research Articles

Formation Obstacle Avoidance: A Fluid-Based Solution

To enhance the mission success rate, the multi-agent system (MAS) often adopts a formation form to shuttle over complex terrain. Compared with obstacle avoidance in other multi-agent coordinated missions, multi-agent formation (MAF) obstacle avoidance is more challenging because agents must both avoid obstacles and subsequently return to their intended positions in the original formation at a uniform speed. Aiming at the three-dimensional complex obstacle environment, the fluid methodology is extended to the application of the formation issue for the first time. In this paper, an MAF obstacle avoidance framework built upon model-based interfered fluid dynamical system (MIFDS), which is the most representative algorithm in fluid methodology, is proposed. The proposed MIFDS is modified to be more in accordance with the kinematic performance of agents than traditional IFDS by introducing the kinematic model and constraints. Moreover, on the basis of MIFDS and the virtual leader formation structure, a novel original fluid speed adjustment strategy in MIFDS, which is the core to solve the formation issue, is designed to balance obstacle avoidance and formation reconfiguration. Finally, receding horizon control is introduced to further guarantee the performance of obstacle avoidance and motion safety. Simulations are used to demonstrate the effectiveness of the proposed method.

Online Bidirectional Trajectory Planning for Mobile Robots in State-Time Space

In this paper, we propose a computationally efficient online local motion planning algorithm for mobile robots in unknown cluttered dynamic environments. The algorithm plans a trajectory incrementally up to the finite horizon in state-time space. Incremental planning method is capable of fast computation but has poor obstacle avoidance performance. To compensate for the drawbacks of incremental planning, a partial trajectory modification scheme is used that sets an interim goal and then plans a trajectory to pass through the interim goal. The interim goal is a temporary desired goal to prevent the robot from falling into the inevitable collision state. By using incremental planning and partial trajectory modification, it is possible to plan collision-free trajectory with small computation even in cluttered dynamic environment. To generate smooth trajectories around given waypoints and goal, we systematize bidirectional trajectory planning with three kinds of trajectories: 1) a forward trajectory from the current robot state; 2) a backward trajectory from the state of current target waypoint; 3) and a connecting trajectory between the forward and backward trajectories. A smooth trajectory is generated around the way point and goal by setting the state of the current target waypoint (or goal), while taking into account the positional relationship among the planned forward trajectory, the target waypoint and the next waypoint. Performances of the proposed algorithm are validated through extensive simulations and experiment with two types of mobile robots: 1) a holonomic mobile robot and 2) a differential drive mobile robot.

Active Virtual Impedance Control for Sound-Following Robots to Avoid Obstacles

A novel active virtual impedance algorithm is here proposed to help sound-following robots avoid obstacles while tracking a sound source. The tracking velocity of a mobile robot to a sound source is determined by virtual repulsive and attraction forces to avoid obstacles and to follow the sound source, respectively. Active virtual impedance is defined as a function of distances and relative velocities to the sound source and obstacles from the mobile robot, which is used to generate the tracking velocity of the mobile robot. Conventional virtual impedance methods have fixed coefficients for relative distances and velocities. However, in this research, the coefficients are dynamically adjusted to extend the obstacle avoidance performance to multiple obstacle environments. The relative distances and velocities are obtained using a microphone array consisting of three microphones in a row. The geometrical relationships of the microphones are utilized to estimate the relative position and orientation of the sound source with respect to the mobile robot, which carries the microphone array. The effectiveness of the proposed algorithm is demonstrated by experiments.

LOW COST COLLISION AVOIDANCE SYSTEM ON HOLONOMIC AND NON- HOLONOMIC MOBILE ROBOTS

Technological advancement in industries necessitates development in autonomous mobile robots with obstacle avoidance for applications in industries, hospitals and educational environment. Holonomic motion allows the mobile robots to move instantaneously in any direction with no constraints while non-holonomic motion restricts the motion in limited directions. This study presents the comparison of obstacle avoidance performance using low cost sensors on holonomic and non-holonomic mobile robots. The robot prototypes are developed and implemented with stable control system for better mobility. Experimental results show that the omnidirectional mobile robots avoid collisions with limited space and less time while non-holonomic mobile platforms exhibits reduced computational complexity and efficient implementation. The developed mobile robots can be used for diverse task specification applications. Article DOI: https://dx.doi.org/10.20319/mijst.2019.51.1222 This work is licensed under the Creative Commons Attribution-Non-commercial 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc/4.0/ or send a letter to Creative Commons, PO Box 1866, Mountain View, CA 94042, USA.

Modified 1D Virtual Force Field Approach to Moving Obstacle Avoidance for Autonomous Ground Vehicles

A modified 1D-virtual force field (1D-VFF) approach is newly proposed to effectively avoid moving obstacles for autonomous ground vehicle (AGV) applications. The underlying idea of this approach is to construct an obstacle force field based on the predicted trajectory of a moving obstacle. To do this, the time-to-collision and the probable area of collision are estimated using the obstacle position and velocity information measured from a FMCW radar. The predicted obstacle force field (POFF) replaces the obstacle force field (OFF) of the conventional 1D-VFF to enhance the AGV’s obstacle avoidance performance. As in the conventional 1D-VFF scheme, the steering force field (SFF) is generated by using the prescribed position of the goal and the current position of the ground vehicle. In order to ensure a collision-free local path against the moving obstacle, the steering command is determined as the maximizing solution of the integrated force field (IFF) defined by a linear combination of POFF and SFF. Simulations of various scenarios show that the proposed algorithm provides an improved local path planning performance against moving obstacles compared to the existing 1D-VFF methodology.

Human Intention Understanding From Multiple Demonstrations and Behavior Generalization in Dynamic Movement Primitives Framework

Human’s interference in the process of skill learning can improve the performance of the robot greatly. However, learning from demonstration to generate a new action with human behavioral characteristics in the varying situation is challenging. Generally, dynamic movement primitives (DMPs) method can generalize the trajectory imitating the demonstration, but cannot integrate the feature of multiple trajectories of different targets. In this paper, the proposed method contains two aspects of learning and generating. The statistical method Gaussian mixture model and Gaussian mixture regression (GMM-GMR) is used to extract the common characteristic and eliminate the uncertainty of the multiple demonstrations. To exert the ability of DMPs to generate a human-like motion to a new goal, and we model the shape parameter with locally weighted regression (LWR) method. To enhance the ability of DMPs in multiple trajectories learning, we propose the multivariate Gaussian process regression (MV-GPR) method to construct the regression model of shape parameters to reflect the human intentions, with respect to the target position. To verify the feasibility of the proposed method, we design a peg-in-hole experiment with proving generalization and obstacle avoidance performance. The results have shown that the strategy integrated the generalization of DMPs and feature regeneration ability of MV-GPR method, and the generated valid trajectory could achieve the peg-in-hole task with 6-DOF whole-arm avoidance.

A 55-nm, 1.0–0.4V, 1.25-pJ/MAC Time-Domain Mixed-Signal Neuromorphic Accelerator With Stochastic Synapses for Reinforcement Learning in Autonomous Mobile Robots

Reinforcement learning (RL) is a bio-mimetic learning approach, where agents can learn about an environment by performing specific tasks without any human supervision. RL is inspired by behavioral psychology, where agents take actions to maximize a cumulative reward. In this paper, we present an RL neuromorphic accelerator capable of performing obstacle avoidance in a mobile robot at the edge of the cloud. We propose an energy-efficient time-domain mixed-signal (TD-MS) computational framework. In TD-MS computation, we demonstrate that the energy to compute is proportional to the importance of the computation. We leverage the unique properties of stochastic networks and recent advances in Q-learning in the proposed RL implementation. The 55-nm test chip implements RL using a three-layered fully connected neural network and consumes a peak power of 690 $\mu \text{W}$ .

Path planning method with obstacle avoidance for manipulators in dynamic environment

Obstacle avoidance is of great importance for path planning of manipulators in dynamic environment. To help manipulators successfully perform tasks, a method of path planning with obstacle avoidance is proposed in this article. It consists of two consecutive phases, namely, collision detection and obstacle-avoidance path planning. The collision detection is realized by establishing point-cloud model and testing intersection of axis-aligned bounding boxes trees, while obstacle-avoidance path planning is achieved through preplanning a global path and adjusting it in real time. This article has the following contributions. The point-cloud model is of high resolution while the speed of collision detection is improved, and collision points can be found exactly. The preplanned global path is optimized based on the improved D-star algorithm, which reduces inflection points and decreases collision probability. The real-time path adjusting strategy satisfies the requirement of reachability and obstacle avoidance for manipulators in dynamic environment. Simulations and experiments are carried out to evaluate the validity of the proposed method, and the method is available to manipulators of any degree of freedom in dynamic environment.

Visual optimization and decision making system for android robot soccer

The paper developed a visual optimization algorithm to ameliorate the problems encountered by androids on the soccer field involving a visual prediction path that can assist the soccer androids in their ball passing function. It includes path planning, obstacle avoidance function, image processing, instant control and smart decision-making techniques. The dynamics of the androids on the soccer field is changing rapidly and the tasks of the soccer androids need to be adjusted according to the actual situation; that means, it needs to transform their roles instantly. Therefore, the system takes decision-tree deduction parameters to make the robot task decision so that the robot can quickly perform obstacle avoidance task and real-time optimization controls on attacking, defending and guarding the goal. According to the result of the experiment, it is evident that the proposed scheme can indeed achieve the role of task allocation and improve the efficiency of motion controls.

Simultaneous Trajectory Planning and Tracking Using an MPC Method for Cyber-Physical Systems: A Case Study of Obstacle Avoidance for an Intelligent Vehicle

As a typical example of cyber-physical systems, intelligent vehicles are receiving increasing attention, and the obstacle avoidance problem for such vehicles has become a hot topic of discussion. This paper presents a simultaneous trajectory planning and tracking controller for use under cruise conditions based on a model predictive control (MPC) approach to address obstacle avoidance for an intelligent vehicle. The reference trajectory is parameterized as a cubic function in time and is determined by the lateral position and velocity of the intelligent vehicle and the velocity and yaw angle of the obstacle vehicle at the start point of the lane change maneuver. Then, the control sequence for the vehicle is incorporated into the expression for the reference trajectory that is used in the MPC optimization problem by treating the lateral velocity of the intelligent vehicle at the end point of the lane change as an intermediate variable. In this way, trajectory planning and tracking are both captured in a single MPC optimization problem. To evaluate the effectiveness of the proposed simultaneous trajectory planning and tracking approach, joint veDYNA-Simulink simulations were conducted in the unconstrained and constrained cases under leftward and rightward lane change conditions. The results illustrate that the proposed MPC-based simultaneous trajectory planning and tracking approach achieves acceptable obstacle avoidance performance for an intelligent vehicle.

End-to-End Autonomous Mobile Robot Navigation with Model-Based System Support

Autonomous mobile robot navigation in real unmodified outdoor areas frequented by people on their business, children playing, fast running bicycles, and even robots, remains a difficult challenge. For eleven years, the Tsukuba Challenge Real World Robot Challenge (RWRC) has brought together robots, researchers, companies, government, and ordinary citizens, under the same outdoor space to push forward the limits of autonomous mobile robots. For the Tsukuba Challenge 2017 participation, our team proposed to study the problem of sensors-to-actuators navigation (also called End-to-End), this is, having the robot to navigate towards the destination on a complex path, not only moving straight but also turning at intersections. End-to-End (E2E) navigation was implemented using a convolutional neural network (CNN): the robot learns how to go straight, turn left, and turn right, using camera images and trajectory data. E2E network training and evaluation was performed at Nagoya University, on similar outdoor conditions to that of Tsukuba Challenge 2017 (TC2017). Even thought E2E was trained on a different environment and conditions, the robot successfully followed the designated trajectory in the TC2017 course. Learning how to follow the road no matter the environment is of the key attributes of E2E based navigation. Our E2E does not perform obstacle avoidance and can be affected by illumination and seasonal changes. Therefore, to improve safety and add fault tolerance measures, we developed an E2E navigation approach with model-based system as backup. The model-based system is based on our open source autonomous vehicle software adapted to use on a mobile robot. In this work we describe our approach, implementation, experiences and main contributions.

Transmission Line Unmanned Aerial Vehicle Obstacle Avoidance System Incorporating Multiple Sensing Technologies

UAVs have been used in a large number of applications in transmission line inspections and job monitoring. Infrared detection perception, ultrasonic radar, visual recognition and other perception technologies are integrated into the obstacle perception of UAVs, adding a variety of obstacle detection methods to the UAV, enabling inductive transmission line equipment without dead angles to be detected early. Line equipment and devices in front of man-machine, and can reasonably perform obstacle avoidance operations, effectively reduce the collision accident between UAVs and lines, and improve the safety and reliability of unmanned aerial vehicles on transmission line inspection work, and ensure transmission lines operational safety has a very important role.

Design and Research of Picking Manipulator Obstacle Avoidance System Based on IOT

In recent years, wireless communication technology has developed rapidly and the speed of wireless communication data transmission is faster and faster, as well as remote control delay has greatly reduced, which is almost negligible, so the remote control is gradually being widely used in various fields. This paper achieves the motion control of picking manipulator and obstacle avoidance system with the advantages of IOT remote control technology and PLC real-time control system. Experiments show that the designed picking manipulator motion control and obstacle avoidance system not only have strong obstacle avoidance ability but also have smooth path optimization with high reliability and stability.

Walking behavior over multiple obstacles in people with Parkinson's disease

The presence of a second obstacle changed the planning and adjustments for obstacle avoidance performance, but this context is poorly understood in Parkinson’s disease (PD). The aim of this study was to investigate the walking behavior over multiple obstacles in people with PD. Nineteen people with PD and 19 healthy individuals walked across an 8m pathway, performing three trials for following conditions: unobstructed walking, walking with one obstacle avoidance (Single), and walking with two obstacles avoidance (Double). In the Double condition, the analysis was performed only for the first obstacle (First Double). The dependent variables were calculated separately for the approach and crossing phases in the obstacle conditions. The main results show that people with PD decreased single support and increased double support phase in both Single and Double conditions compared to the unobstructed walking. Both groups increased stride duration during approach phase in the Double condition compared to the unobstructed walking and Single conditions. The presence of the second obstacle led to a decrease in trailing toe clearance during obstacle avoidance of the First Double. In conclusion, people with PD use a conservative strategy while approaching obstacles. Both groups need more time to obtain and process environmental information and plan the action in environments with multiple obstacles. The smaller leading toe clearance might be an indicative that the presence of a second obstacle increase the likelihood of tripping during obstacle avoidance in both people with PD and healthy individuals.

Obstacle Avoidance and Secondary Task Performance During Locomotion

Obstacle Avoidance and Secondary Task Performance During Locomotion

Dual tasking negatively impacts obstacle avoidance abilities in post-stroke individuals with visuospatial neglect: Task complexity matters!

Persons with perceptual-attentional deficits due to visuospatial neglect (VSN) after a stroke are at a risk of collisions while walking in the presence of moving obstacles. The attentional burden of performing a dual-task may further compromise their obstacle avoidance performance, putting them at a greater risk of collisions. The objective of this study was to compare the ability of persons with (VSN+) and without VSN (VSN-) to dual task while negotiating moving obstacles. Twenty-six stroke survivors (13 VSN+, 13 VSN-) were assessed on their ability to (a) negotiate moving obstacles while walking (locomotor single task); (b) perform a pitch-discrimination task (cognitive single task) and (c) simultaneously perform the walking and cognitive tasks (dual task). We compared the groups on locomotor (collision rates, minimum distance from obstacle and onset of strategies) and cognitive (error rates) outcomes. For both single and dual task walking, VSN+ individuals showed higher collision rates compared to VSN- individuals. Dual tasking caused deterioration of locomotor (more collisions, delayed onset and smaller minimum distances) and cognitive performances (higher error rate) in VSN+ individuals. Contrastingly, VSN- individuals maintained collision rates, increased minimum distance, but showed more cognitive errors, prioritizing their locomotor performance. Individuals with VSN demonstrate cognitive-locomotor interference under dual task conditions, which could severely compromise safety when ambulating in community environments and may explain the poor recovery of independent community ambulation in these individuals.

Dynamic obstacle avoidance for manipulators using distance calculation and discrete detection

Abstract In order to avoid dynamic obstacle timely during manufacturing tasks performed by manipulators, a novel method based on distance calculation and discrete detection is proposed. The nearest distances between the links of a manipulator and the convex hull of an arbitrarily-shaped dynamic obstacle obtained from Kinect-V2 camera in real-time are calculated by Gilbert–Johnson–Keerthi algorithm, and the minimum one is defined as the closest distance between the manipulator and the obstacle. When the closest distance is less than a safe value, whether the dynamic obstacle is located in the global path of the manipulator is determined by improved discrete collision detection, which can adjust detection step-size adaptively for accuracy and efficiency. If the obstacle will collide with the manipulator, set a local goal and re-plan a local path for the manipulator. The proposed method is implemented in Robot Operating System (ROS) using C++. The experiments indicate that the proposed method can perform safe and timely dynamic avoidance for redundant manipulators in human-robot interaction.

An Active Virtual Impedance Control Algorithm for Collision Free Navigation of a Mobile a Robot

<span style="color: black; font-family: 'Times New Roman','serif'; font-size: 9pt; mso-fareast-font-family: '맑은 고딕'; mso-bidi-font-style: italic; mso-fareast-theme-font: minor-fareast; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;" lang="EN-US">An modified active virtual impedance control has been proposed for collision free navigation of a mobile robot to avoid front obstacles dynamically while a mobile robot is following a sound source. A mobile robot is controlled to follow a sound source with a velocity which is determined by virtual repulsive and attraction forces to avoid obstacles and to follow the sound source, respectively. To generate the virtual repulsive and attraction forces, a new modified virtual impedance is defined as a function of the distances and relative velocities to the sound source and obstacles from the mobile robot. In the conventional virtual impedance method, fixed coefficients have been used for the virtual impedance control. In this research, the coefficients are dynamically adjusted to elaborate the obstacle avoidance performance in various situations such as the multiple moving obstacles environment. A microphone array consisting of three microphones in a row has been attached on the mobile robot to detect the relative distance and velocity to the obstacles. The relative position and orientation of the sound source against the mobile robot has been estimated using the geometrical relationship of the microphones. As an application, the mobile robot can be used as a pet robot following the master with a sound source. The effectiveness of the proposed algorithm has been demonstrated through real experiments.</span>

Robust analysis of an underwater navigational strategy in electrically heterogeneous corridors

Obstacles and other global stimuli provide relevant navigational cues to a weakly electric fish. In this work, robust analysis of a control strategy based on electrolocation for performing obstacle avoidance in electrically heterogeneous corridors is presented and validated. Static output feedback control is shown to achieve the desired goal of reflexive obstacle avoidance in such environments in simulation and experimentation. The proposed approach is computationally inexpensive and readily implementable on a small scale underwater vehicle, making underwater autonomous navigation feasible in real-time.

Cues and Attention in Parkinsonian Gait: Potential Mechanisms and Future Directions.

Cues and Attention in Parkinsonian Gait: Potential Mechanisms and Future Directions.